|Publication number||US6931568 B2|
|Application number||US 10/113,167|
|Publication date||Aug 16, 2005|
|Filing date||Mar 29, 2002|
|Priority date||Mar 29, 2002|
|Also published as||US20030188222|
|Publication number||10113167, 113167, US 6931568 B2, US 6931568B2, US-B2-6931568, US6931568 B2, US6931568B2|
|Inventors||Antonio Abbondanzio, Gregory William Dake, Dhruv Manmohandas Desai, Benjamin Russell Grimes, William Joseph Piazza, Gregory Brian Pruett, David B. Rhoades|
|Original Assignee||International Business Machines Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (23), Non-Patent Citations (1), Referenced by (60), Classifications (7), Legal Events (6)|
|External Links: USPTO, USPTO Assignment, Espacenet|
1. Field of the Present Invention
The present invention generally relates to the field of data processing systems and more particularly to a server system employing redundant service processors and to a method for managing the service processors when one fails.
2. History of Related Art
The concept of a service processor is well know in the field of data processing systems and particularly server class systems. Service processors are provided to manage certain aspects of a server and can be defined loosely as an auxiliary processor that monitors the environment and health of one or more main processors and their associated subsystems. In the event of an actual or predicted problem with a main processor, subsystem, or the environment, the service processor is capable of taking action to alert a system administrator and/or to correct the problem on its own. Service processors have been used on large systems such as IBM mainframes for decades and are now well established as a management tool for industry standard class servers including servers based on x86 processors such as the PentiumŪ family of processors from Intel.
Redundant service processors may be provided in high-availability systems so that a failure of one service processor does not result in the loss of the monitoring, alerting, and initialization capabilities that are imperative in such systems. Implementing redundant service processors is complicated for a number of reasons. A method by which the service processors agree which one is in control must be defined and subscribed to by both service processors. All subsystems, including the service processors themselves, must understand which service processor is in control at any given point in time. In some instances, hardware must be provided to switch busses from one service processor to the other and this hardware must switch the busses synchronously with fail-over of control from one service processor to the other. For purpose of this disclosure “fail-over” refers to a transfer of control from one service processor to another.
There are today a number of methods used in the industry to provide fail-over of redundant subsystems and to coordinate the actions of redundant controllers. Network adapter cards, for example, are sometimes installed in server systems as a redundant pair and a device driver is given the responsibility of transferring traffic from one to the other in the event of a failure of the active card. Those skilled in the art may also be familiar with “voting” systems in which an odd number of redundant controllers make a decision independently but simultaneously and the action taken depends upon what the majority of the systems agree is the proper course of action. Such systems, unfortunately, are typically complex, as evidenced by the delay in the first space shuttle launch caused by a failure of multiple on-board computers to synchronize their communications with each other in such a redundant-controller environment.
It would be desirable to implement a method and system that enables the use of redundant service processors and implements all related control issues. It would be further desirable if the implemented solution did not substantially increase the cost or complexity of the system.
The problems identified above are in large part addressed by a system and method for determining an active service processor from two or more redundant service processors in the system. The system includes at least one managed subsystem represented by, for example, a processor planar (blade), two (or possibly more) management modules, and an interconnection plane. The processor blade (server blade) includes one or more general purpose microprocessors, memory, and I/O devices according to well known microprocessor-based design principles. Each management module includes a service processor and control logic connected to the service processor. The control logic is configured to receive various status signals and to generate a control signal based thereon. The control signal is provided, via the interconnect plane, to determination logic on each managed subsystem. The determination logic receives a control signal from each management module and generates a switch signal based on the state of the control signals. The switch signal controls switching logic configured to receive bus signals from the service processors on each management module. Based on the control signal, one of the service processor bus signals is provided to managed instrumentation on the managed subsystem. The management module control logic is generally configured to maintain the control signal in its current state if the active processor is determined to be functional. The control logic is further configured to alter the control signal state if the active service processor is determined to be non-functional. A transition in the control signal typically generates a fail-over event that causes the switching logic on the managed subsystems to switch from the previously active service processor to the previously inactive or standby service processor as the source of service processor signals. The management module may further include a watchdog timer that is used to reset the local service processor if the processor “hangs.”
Other objects and advantages of the invention will become apparent upon reading the following detailed description and upon reference to the accompanying drawings in which:
While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the drawings and detailed description presented herein are not intended to limit the invention to the particular embodiment disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the present invention as defined by the appended claims.
Generally speaking the invention contemplates a data processing system implementing redundant service processors and a corresponding method for handling fail-over control in the system. The system typically includes two management modules and at least one managed subsystem such as a server blade. Each management module includes a service processor and control logic configured to receive various status signals from the service processor and to generate a control signal based thereon. The fail-over mechanism must be able to handle various fail-over scenarios. A first fail-over scenario occurs when one of the service processors is not installed and the remaining service processor must assume control after initialization. Another scenario occurs when one of the service processors becomes non-responsive (i.e., “hangs”), either while initializing itself, while acting as a standby for the other service processor, or while acting as the active service processor in control of the system. A third scenario occurs when one of the service processors detects an internal failure and either “surrenders” control to the other service processor or takes itself out of consideration for becoming the active service processor in the future. A fourth fail-over scenario occurs when the standby service processor detects a failure of the active service processor and usurps control of all subsystems from the active service processor. The implemented system is preferably capable of handling each of these scenarios.
Before describing specific features of the management modules that include the redundant service processors and implement the fail-over mechanism according to the present invention, selected elements of a data processing system typical of the subsystems managed by the management modules are described. Turning to the drawings,
In server blade 100, a bus bridge 108 provides an interface between system bus 104 and an I/O bus 110 to which one or more peripheral devices 114A through 114N (generically or collectively referred to as peripheral device(s) 114) as well as a general purpose I/O (GPIO) port are connected. Peripheral devices 114 may include devices such as a graphics adapter, a high-speed network adapter or network interface card (NIC), a hard-disk controller, and the like. I/O bus 110 is typically compliant with one of several industry standard I/O bus specifications including, as a common example, the Peripheral Components Interface (PCI) bus as specified in PCI Local Bus Specification Rev 2.2 by the PCI Special Interest Group (www.pcisig.com).
The depicted embodiment of server blade 100 includes a subsystem service processor 116 connected to GPIO port 112. Subsystem service processor 116 is configured to provide support for main processors 102. This support may include, for example, monitoring the power supplied to main processor(s) 102 and, in the event of a blade crash, initiating a restart of the main processors. Although each server blade 100 may include a subsystem service processor 116, the present invention is primarily associated with redundant service processors on a management blade associated with a collection or network of service blades as described in greater detail below.
Turning now to
As shown in the front view of
The number of server blades within cabinet 201 varies with the implementation. In a representative configuration, the front face 203 of cabinet 201 includes 14 or more slots 202 for receiving server blades 100. Each server blade 100 is typically a full-height adapter.
The rear view of data processing configuration 200 depicted in
Generally speaking, each management module 120 is configured to monitor and control resources and characteristics of subsystems in data processing configuration 200 that are shared by each server blade 100. These resources and characteristics may include, for example, the power applied to cabinet 201, cabinet cooling fans, and environmental characteristics such as the ambient temperature within cabinet 201.
Portions of the present invention may be implemented as a sequence of processor executable instructions (software) for monitoring the insertion and removal of server blades in a data processing network where the instructions are stored on a computer readable medium. During execution, portions of the software may reside in a volatile storage element such as the memory 141 (
Turning now to
In the depicted embodiment, management module 120 a includes fail-over control logic (FOCL) 132 a and a watchdog timer 134 a. FOCL 132 a is typically configured to allow its corresponding service processor 130 a to control a fail-over control signal (FOC 0) 131 a. Fail-over control signal 131 a, in conjunction with a corresponding signal identified as fail-over control signal (FOC 1) 131 b generated by service module 130 b, are used to initiate and indicate a change in the identity of the active service processor. The two FOC signals 131 can indicate one of four possible combinations. In one embodiment, the identity of the active service processor is determined based upon the parity of the FOC signals 131.
Service modules 120 manage subsystems of configuration 200. In
AMDL logic 142 generates a switch control signal 143 that is asserted whenever there is a change in the identity of the active service processor. Switch control signals 143 provides an input to bus switch logic 140. In addition, switch control signal 143 may be provided to a processor on the managed subsystem to enable the subsystem to determine the identity of the active processor. In the case where the managed subsystem is a server blade 100, switch control signal 143 may be provided either to the blade's general purpose (main) processor 102 or to its subsystem service processor 116 depending upon the implementation.
In the depicted embodiment, each service processor 130 generates management logic signals that are provided to managed subsystems such as server blade 100 to control or monitor various aspects of each managed subsystem. Typically, these management logic signals, represented by reference numerals 133 a and 133 b (generically or collectively referred to as management logic signal(s) 133), comply with an industry standard signal bus protocol or specification. Management logic signals 133 may comply, for example, with the RS-232, RS-485, or I2C specifications. Management logic signals 133 from each service processor 130 are provided to bus switch logic 140 on server blade 100. Bus switch logic 140 is configured as a switching network that connects the managed subsystem to the management control signals of the active processor based upon the state of switch control signal 143. Thus, bus switch logic 140 acts as a bus-width multiplexer that provides the management control signals from the active service processor to managed instrumentation, represented by reference numeral 144, within the managed subsystem. Management instrumentation 144 represents devices or logic within the managed subsystem that are configured to respond to the active service processor. Managed instrumentation 144 generally identifies that portion of the managed subsystem that provides visibility to the status and health of the subsystem and provides control access to the subsystem.
As further depicted in
Each management module 120 as depicted in
Management modules 120 may further include a watchdog timer identified by reference numeral 134. Watchdog timers are well known and commonly used in the industry to detect a non-responsive processor and reset it. Watchdog timer 134 receives a watchdog signal 136 from service processor 130. Watchdog signal 136 informs watchdog timer 134 that the service processor is alive. In one embodiment, service processor 130 will assert watchdog timer signal 136 periodically at some programmable or predeterminable interval under normal operating conditions. If watchdog signal 136 is not asserted within the specified predetermined interval, watchdog timer 134 will assert a reset signal 139 that is provided to appropriate reset inputs of service processor 130 and FOCL 132.
In the embodiment depicted in
From a functional perspective, the depicted embodiment of the present invention operates as follows. (The following description is from the perspective of management module 120 a such that “local” refers to logic or signals on module 120 a while “remote” refers to logic or signals on module 120 b). FOCL 132 on each management module 120 produces an FOC signal 131 that is routed through interconnection plane 145 to each server blade 100 and to the other management module. AMDL 142 on each server blade combines the two FOC signals 131 into a switch control signal 143. AMDL 142 may be implemented as an XOR gate. Weak pull-up resistors on each server blade 100 connected to FOC signals 131 assure that the signals assume a known and stable state even if one or both management modules 120 are not installed.
FOCL 132 functions by evaluating status signals 135 produced by service processor 130 and generating FOC signal 131 based thereon. FOC signal 131 can be used to initiate a fail-over of service processors under appropriate conditions. In a typical embodiment, status signals 135 include a local OK (LOK#) signal 152 (see
In addition, status signals 135 may include a ready signal 151. The state of ready signal 151 indicates that the corresponding service processor 130 has completed all local and remote assessments and is ready to participate in the active service processor determination process. If ready signal 151 is not asserted, FOC signal 131 assumes a high-impedance state as if its management module 120 were not installed. After a hardware reset, ready signal 151 is initialized to the state that disables FOC signal 131 (i.e., the reset signal must set FOC signal 131 to a high-Z state after reset).
If the watchdog timer 134 is included, the status signals 135 may further include a watchdog timer reset signal 139 that indicates, when asserted, that watchdog timer 134 was allowed to expire and that, therefore, the local service processor 130 appears non-responsive and unable to perform duties as the active service processor. After hardware reset, the state of this signal is not critical because the ready line will prevent any values of this signal from affecting the determination of the active manager. It is assumed that watchdog timer 134 allows enough time after reset for its service processor 130 to initialize and begin to manage the watchdog timer before the timer resets the service processor again.
In addition to these status signals, FOCL 132 typically receives a clock signal (CLK) and a power good reset signal (RST#). The clock signal determines the appropriate times for FOC signal 131 to transition. It is not necessary that the clock signal for management module 120 a be synchronized to the clock signal of management module 120 b. The power good reset signal typically initializes all logic to a known state when power is applied to management module 120.
In one embodiment, FOCL 132 combines the signals indicated in the preceding to produce FOC signal 131 in compliance with the following guidelines. FOC 131 must assume a high-impedance state if the local service processor 130 is not ready to participate in the active manager determination process. FOC signal 131 maintains its current state (whether high or low) if either of two conditions are TRUE. The first condition is determined by the combination of the local FOC signal 131 a and the remote FOC signal 131 b, and the (local) position signal 137 a. If these signals indicate that service processor 130 a is in control, that service processor 130 a is functioning, and that service processor 130 a does not have a watchdog timer reset pending, the state of local control signal 131 a is maintained. Similarly, if these signals indicate that the remote service processor 130 b is the active service processor and that the local service processor 130 a has determined that the active service processor (remote service processor 130 b) is healthy, no state change of FOC signal 131 a is made. Thus, FOC signal 131 will not transition if the active service processor 130 is determined to be fully functioning whether the active processor is local or remote.
On the other hand, local FOC signal 131 a will transition (change state) when either of two conditions is present. The first condition (for transitioning FOC signal 131) is met when the combination of local FOC signal 131 a, remote FOC signal 131 b, and local position signal 137 a, indicate that the local service processor 130 a is the active service processor and that the service processor is either non-functional or has a watchdog time reset pending. The second condition for toggling FOC signal 131 a is met if the status signals indicate that remote service processor 130 b is the active service processor and that local service processor 130 a has determined that the remote service proper is non-functional.
Referring now to
In addition to the hardware illustrated in the figures and described in the preceding paragraphs, management modules 120 typically include software that facilitates the determination of the active service processor. More specifically, the management module software is preferably configured to implement the following methods of initializing and managing FOCL 132. Referring to
Referring now to
Local service processor 130 a will also attempt periodically to determine (block 194) the state of remote service processor 130 b. If inter-processor communications are possible via multiple channels, the preferred embodiment attempts to communicate over two or more of the available channels. Whenever an inter-processor communication link is established where none previously existed or all previously established links are lost, local service processor 130 a must change the state of the ROK# signal to reflect the new status. If (block 195) the new status of remote service processor is not healthy, the ROK# signal is de-asserted (block 196). The de-assertion of ROK# will sometimes cause FOCL 132 a to transition FOC signal 131 a if is appropriate to do so. More specifically, if FOCL 132 a senses the ROK# signal 153 a change to a healthy status, no action is taken because it is not necessary to switch control when the remote processor comes on-line. If ROK# signal 131 a transitions to an unhealthy state, FOCL 132 a will act depending upon the identity of the currently active service processor. If FOCL 132 a determines (block 197) the remote service to be the active processor based on the state of FOC 131 a, FOC 131 b, and TOP# 137 a, FOCL 132 a will toggle (block 198) its FOC signal 131 a to force control to pass to the local service processor.
Local service processor 130 a is also configured to track (block 250) fail-over events that occur within a specified time period. If the number of fail-over events that occur during the specified interval exceed a predetermined maximum (block 252), the service processor is configured to declare itself ineligible to participate in the service processor determination process. A service processor declares itself ineligible by de-asserting (block 254) its READY signal 151. This mechanism prevents a situation in which each service processor detects that the other is non-functional and attempts to gain control by initiating a fail-over. When the fail-over is complete, the non-active processor will determine that the new active processor is non-functional and repeat the process indefinitely. By placing a limit on the number of fail-over events that can occur in a specified interval, the invention prevents an endless loop scenario in which neither service processor is in control. Generally speaking, fail-overs are monitored by interrupting the service processor each time a fail-over occurs, counting the number of interrupts, and de-asserting the READY signal if the number of fail-over events exceeds the specified threshold.
Following a hardware reset, the selection of which service processor 130 will act as the initial active processor is non-deterministic. Because management modules 120 are configured as substantially identical and redundant elements, however, the outcome of the initial service processor determination following reset is irrelevant. Initially, the switch control signals 143 of each managed subsystem will favor one of the service processors because the FOC signals 131 will be floating and the pull-up resistors will bias the switch signal to favor one of the service processors. After a service processor 130 fails, the amount of time that passes before a fail-over occurs is dependent upon the duration of the watchdog timer interval, the speed of the clock to the fail-over control logic, and the frequency with which the other service processor checks for communications between the management modules. Careful selection of these parameters can achieve a fail-over response time of a few seconds down to less than a second.
Both management modules 120 a and 120 b are capable of presenting themselves on local area configuration 200 and must, therefore, have a network address such as an IP address, MAC address, or other suitable and unique identifier. If an application program such as a management and deployment application exemplified by the IBM Director program is attempting to communicate with the active management module at the time the management module fails, the preferred embodiment of the invention migrates the IP address from the hung service processor to the new active service processor as part of the fail-over sequence. This process is facilitated by configuring each management module 120 to know which module is in control at any time. During initialization, the active management module 120 may pass its identifying address to the remote management module. When a fail-over occurs, the active module stops responding to the IP address and the standby module must program its network stack to begin responding to requests set to that address. Network traffic in transit when a fail-over occurs will either be responded to by the new active management module or will time out. In the case of a time-out, a subsequent retry will likely be successful.
It will be apparent to those skilled in the art having the benefit of this disclosure that the present invention contemplates a system and method for controlling the active service processor in a system featuring redundant service processors. It is understood that the form of the invention shown and described in the detailed description and the drawings are to be taken merely as presently preferred examples. It is intended that the following claims be interpreted broadly to embrace all the variations of the preferred embodiments disclosed.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US3786433 *||Sep 25, 1972||Jan 15, 1974||Kent Ltd G||Computer control arrangements|
|US4153198 *||Feb 1, 1977||May 8, 1979||Hitachi, Ltd.||Electro-hydraulic governor employing duplex digital controller system|
|US4350225 *||Jan 31, 1980||Sep 21, 1982||Hitachi, Ltd.||Elevator control system|
|US4351023 *||Apr 11, 1980||Sep 21, 1982||The Foxboro Company||Process control system with improved system security features|
|US4562575||Jul 7, 1983||Dec 31, 1985||Motorola, Inc.||Method and apparatus for the selection of redundant system modules|
|US5185693||Jan 29, 1992||Feb 9, 1993||Olin Corporation||Method and apparatus for providing backup process control|
|US5251299||Mar 19, 1990||Oct 5, 1993||Fujitsu Limited||System for switching between processors in a multiprocessor system|
|US5434998||Mar 16, 1993||Jul 18, 1995||Yokogawa Electric Corporation||Dual computer system|
|US5696895||Jun 19, 1995||Dec 9, 1997||Compaq Computer Corporation||Fault tolerant multiple network servers|
|US5781715||Oct 13, 1992||Jul 14, 1998||International Business Machines Corporation||Fault-tolerant bridge/router with a distributed switch-over mechanism|
|US5784551 *||Sep 15, 1993||Jul 21, 1998||Siemens Telecommunicazioni S.P.A.||Duplicate control and processing unit for telecommunications equipment|
|US5898829 *||Nov 27, 1996||Apr 27, 1999||Nec Corporation||Fault-tolerant computer system capable of preventing acquisition of an input/output information path by a processor in which a failure occurs|
|US6002970 *||Oct 15, 1997||Dec 14, 1999||International Business Machines Corp.||Method and apparatus for interface dual modular redundancy|
|US6065135 *||Jun 4, 1999||May 16, 2000||Lockhead Martin Corporation||Error detection and fault isolation for lockstep processor systems|
|US6138247||May 14, 1998||Oct 24, 2000||Motorola, Inc.||Method for switching between multiple system processors|
|US6240526||Oct 27, 1999||May 29, 2001||Resilience Corporation||Triple modular redundant computer system|
|US6425092 *||Jun 17, 1998||Jul 23, 2002||International Business Machines Corporation||Method and apparatus for preventing thermal failure in a semiconductor device through redundancy|
|US6470462 *||Feb 25, 1999||Oct 22, 2002||Telefonaktiebolaget Lm Ericsson (Publ)||Simultaneous resynchronization by command for state machines in redundant systems|
|US6550018 *||Feb 18, 2000||Apr 15, 2003||The University Of Akron||Hybrid multiple redundant computer system|
|US6785840 *||Aug 31, 2000||Aug 31, 2004||Nortel Networks Limited||Call processor system and methods|
|US20030159084 *||Jan 10, 2000||Aug 21, 2003||Sun Microsystems, Inc.||Controlled take over of services by remaining nodes of clustered computing system|
|JPH07160522A||Title not available|
|JPS58225401A||Title not available|
|1||Selection Mechanism for Active vs. Backup Processor; IBM Technical Disclosure Bulletin, 12-95, pp. 75-78, vol. 38, No. 12.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7085961 *||Dec 30, 2002||Aug 1, 2006||Quanta Computer Inc.||Redundant management board blade server management system|
|US7114095 *||May 31, 2002||Sep 26, 2006||Hewlett-Packard Development Company, Lp.||Apparatus and methods for switching hardware operation configurations|
|US7171524 *||May 3, 2004||Jan 30, 2007||Hitachi, Ltd.||Storage control system and control method for storage control which suppress the amount of power consumed by the storage control system|
|US7251746 *||Jan 21, 2004||Jul 31, 2007||International Business Machines Corporation||Autonomous fail-over to hot-spare processor using SMI|
|US7260741||Sep 18, 2001||Aug 21, 2007||Cedar Point Communications, Inc.||Method and system to detect software faults|
|US7343436 *||Oct 29, 2002||Mar 11, 2008||Kabushiki Kaisha Forks||Synchronous electronic control system and system control method|
|US7398293 *||Apr 17, 2002||Jul 8, 2008||Dell Products L.P.||System and method for using a shared bus for video communications|
|US7480725||Sep 16, 2004||Jan 20, 2009||Invensys Systems, Inc.||Transparent relocation of an active redundant engine in supervisory process control data acquisition systems|
|US7536493||Apr 12, 2006||May 19, 2009||International Business Machines Corporation||Method and apparatus for identifying a service processor with current setting information|
|US7536604 *||Mar 14, 2006||May 19, 2009||International Business Machines Corporation||Method and system for reconfiguring functional capabilities in a data processing system with dormant resources|
|US7587523 *||Dec 2, 2003||Sep 8, 2009||Cedar Point Communications, Inc.||Distributed systems for determining card status|
|US7593320 *||Apr 30, 2004||Sep 22, 2009||Marvell International, Ltd.||Failover scheme for stackable network switches|
|US7733783||Dec 2, 2003||Jun 8, 2010||Cedar Point Communications, Inc.||Ethernet network availability|
|US7751310 *||Dec 20, 2005||Jul 6, 2010||Nec Corporation||Fault tolerant duplex computer system and its control method|
|US7752493 *||Jul 16, 2004||Jul 6, 2010||Fujitsu Limited||High reliability system, redundant construction control method, and program|
|US7809995 *||Dec 16, 2005||Oct 5, 2010||Siemens Aktiengesellschaft||Method and device for monitoring the functionality of an automation system of a plant|
|US7818615 *||Sep 16, 2004||Oct 19, 2010||Invensys Systems, Inc.||Runtime failure management of redundantly deployed hosts of a supervisory process control data acquisition facility|
|US7836335 *||Apr 11, 2008||Nov 16, 2010||International Business Machines Corporation||Cost-reduced redundant service processor configuration|
|US7870417 *||Apr 20, 2007||Jan 11, 2011||International Business Machines Corporation||Apparatus, system, and method for adapter card failover|
|US8037364||Jan 9, 2009||Oct 11, 2011||International Business Machines Corporation||Forced management module failover by BMC impeachment consensus|
|US8661290 *||Jan 14, 2011||Feb 25, 2014||International Business Machines Corporation||Saving power in computing systems with redundant service processors|
|US8707290 *||Feb 22, 2006||Apr 22, 2014||Dell Products L.P.||Firmware update in an information handling system employing redundant management modules|
|US9146705 *||Apr 9, 2012||Sep 29, 2015||Microsoft Technology, LLC||Split brain protection in computer clusters|
|US9189314 *||Dec 23, 2013||Nov 17, 2015||Hong Fu Jin Precision Industry (Shenzhen) Co., Ltd.||Electronic device and method for detecting firmware of BMC|
|US20030056152 *||Sep 18, 2001||Mar 20, 2003||Fitzgerald Jeffrey J.||Method and system to detect software faults|
|US20030200345 *||Apr 17, 2002||Oct 23, 2003||Dell Products L.P.||System and method for using a shared bus for video communications|
|US20030226057 *||May 31, 2002||Dec 4, 2003||Milojicic Dejan S.||Apparatus and methods for switching hardware operation configurations|
|US20040153697 *||Dec 30, 2002||Aug 5, 2004||Ying-Che Chang||Blade server management system|
|US20050018612 *||Dec 2, 2003||Jan 27, 2005||Fitzgerald Jeffrey J.||Ethernet network availability|
|US20050028021 *||Dec 2, 2003||Feb 3, 2005||English Daniel W.||Distributed systems for determining card status|
|US20050060427 *||Aug 9, 2004||Mar 17, 2005||Sun Microsystems, Inc.||Object-aware transport-layer network processing engine|
|US20050172164 *||Jan 21, 2004||Aug 4, 2005||International Business Machines Corporation||Autonomous fail-over to hot-spare processor using SMI|
|US20050197718 *||Jul 16, 2004||Sep 8, 2005||Fujitsu Limited||High reliability system, redundant construction control method, and program|
|US20050210191 *||May 3, 2004||Sep 22, 2005||Naotaka Kobayashi||Storage control system and control method for storage control system|
|US20050257654 *||Mar 21, 2005||Nov 24, 2005||Jen-Hsuen Huang||Blade server system|
|US20060041700 *||Oct 29, 2002||Feb 23, 2006||Kabushiki Kaisha Forks||System controller, control system, and system control method|
|US20060056285 *||Sep 16, 2004||Mar 16, 2006||Krajewski John J Iii||Configuring redundancy in a supervisory process control system|
|US20060059478 *||Sep 16, 2004||Mar 16, 2006||Krajewski John J Iii||Transparent relocation of an active redundant engine in supervisory process control data acquisition systems|
|US20060069946 *||Sep 16, 2004||Mar 30, 2006||Krajewski John J Iii||Runtime failure management of redundantly deployed hosts of a supervisory process control data acquisition facility|
|US20060133410 *||Dec 20, 2005||Jun 22, 2006||Nec Corporation||Fault tolerant duplex computer system and its control method|
|US20070011407 *||Mar 14, 2006||Jan 11, 2007||Klaus-Juergen Kuehl||Method and system for reconfiguring functional capabilities in a data processing system with dormant resources|
|US20070101059 *||Dec 13, 2006||May 3, 2007||Naotaka Kobayashi||Storage control system and control method for storage control which suppress the amount of power consumed by the storage control system|
|US20070150713 *||Dec 22, 2005||Jun 28, 2007||International Business Machines Corporation||Methods and arrangements to dynamically modify the number of active processors in a multi-node system|
|US20070198875 *||Mar 24, 2004||Aug 23, 2007||Jung Yeong W||Method for trunk line duplexing protection using a hardware watchdog|
|US20070220120 *||Jan 28, 2005||Sep 20, 2007||Takashi Tsunehiro||Computer System|
|US20070234332 *||Feb 22, 2006||Oct 4, 2007||Dell Products L.P.||Firmware update in an information handling system employing redundant management modules|
|US20070245406 *||Apr 12, 2006||Oct 18, 2007||Anderson Gary D||Method and apparatus for identifying a service processor with current setting information|
|US20080126854 *||Sep 27, 2006||May 29, 2008||Anderson Gary D||Redundant service processor failover protocol|
|US20080263255 *||Apr 20, 2007||Oct 23, 2008||International Business Machines Corporation||Apparatus, System, and Method For Adapter Card Failover|
|US20080263391 *||Apr 20, 2007||Oct 23, 2008||International Business Machines Corporation||Apparatus, System, and Method For Adapter Card Failover|
|US20090100292 *||Dec 16, 2005||Apr 16, 2009||Dieter Kleyer||Method and Device for Monitoring the Functionality of an Automation System of a Plant|
|US20090259884 *||Apr 11, 2008||Oct 15, 2009||International Business Machines Corporation||Cost-reduced redundant service processor configuration|
|US20100180161 *||Jul 15, 2010||International Business Machines Corporation||Forced management module failover by bmc impeachment concensus|
|US20120185726 *||Jan 14, 2011||Jul 19, 2012||International Business Machines Corporation||Saving Power in Computing Systems with Redundant Service Processors|
|US20130268495 *||Apr 9, 2012||Oct 10, 2013||Microsoft Corporation||Split brain protection in computer clusters|
|US20130339566 *||Aug 21, 2013||Dec 19, 2013||Fujitsu Limited||Information processing device|
|US20150052381 *||Dec 23, 2013||Feb 19, 2015||Hon Hai Precision Industry Co., Ltd.||Electronic device and method for detecting firmware of bmc|
|US20150168993 *||Dec 27, 2013||Jun 18, 2015||Emerson Network Power - Embedded Computing, Inc.||Safety Relay Box System|
|CN101065731B||Sep 13, 2005||Apr 27, 2011||因文西斯系统公司||System and method for runtime of redundant host pair for processing control network environment|
|WO2006033882A2 *||Sep 13, 2005||Mar 30, 2006||Invensys Sys Inc||Runtime failure management of redundantly deployed hosts of a supervisory process control data acquisition facility|
|Cooperative Classification||H04L1/22, G06F11/2028, G06F11/2038|
|European Classification||G06F11/20P2E, H04L1/22|
|Mar 29, 2002||AS||Assignment|
Owner name: INTERNATIONAL BUSINESS MACHINES CORP., NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ABBONDANZIO, ANTONIO;DAKE, GREGORY W.;DESAI, DHRUV M.;AND OTHERS;REEL/FRAME:012753/0919;SIGNING DATES FROM 20020325 TO 20020327
|Feb 17, 2009||FPAY||Fee payment|
Year of fee payment: 4
|Feb 23, 2009||REMI||Maintenance fee reminder mailed|
|Apr 1, 2013||REMI||Maintenance fee reminder mailed|
|Jun 4, 2013||SULP||Surcharge for late payment|
Year of fee payment: 7
|Jun 4, 2013||FPAY||Fee payment|
Year of fee payment: 8